We have been studying crystallins, the abundant water-soluble proteins responsible for the optical properties of the eye lens, as a model for tissue-specific gene expression. Crystallins are especially interesting since they are also often expressed in non- lens tissues at lower concentrations where they have enzymatic or other non-refractive functions. Within this fiscal year we have published a trabecular meshwork (TM) cell line from mice that can be used to study the expression of transfected genes. This was originally developed to study the regulation of alphaB-crystallin promoter activity, which we showed earlier occurs in these cells. We have also submitted a manuscript showing that myocilin mRNA is present in human intact and organ-cultured TM, but not in human primary TM cell cultures. Myocilin mRNA could be induced from the TM cell cultures by TGFbeta, dexamethasone or mechanical stretch. We have completed the publication initiated last year showing that retinoblastoma protein (pRB) and the TATA box-binding protein (TBP) interact with PAX-6 via its homeodomain, suggesting overlapping pathways for Pax-6 during lens fiber cell differentiation. We have identified a cis-control sequence (PE2A) operating via an AP-1- mediated mechanism in the non-translated sequence of exon 1 of the mouse alphaA-crystallin gene. This suggests previously unrecognized similarities in the mechanism of expression of the mouse and chicken alphaA-crystallin genes and is consistent with the idea that a redox-sensitive mechanism is a selective force for recruiting lens crystallins. We have also submitted a manuscript demonstrating that various lens-specific transgenes encoding trauncated Pax-6 proteins lacking their activation domains act in a dominant negative fashion and induce cataracts in the transgenic mice. Several years ago we showed that both the alphaA and alphaB-crystallins had the ability to undergo limited serine-specific, magnesium-dependent, cyclic AMP- independent autophsphorylation. We now have shown that serine- specific, cyclic AMP-independent autophosphorylation extends to betaB2-crystallin; this polypeptide shares with alpha-crystallins the properties of being also phosphorylated by a cyclic AMP- dependent process and being expressed outside of the lens. Last year we reviewed crystallin phosphorylation and suggest that these crystallins my be involved in a yet to be discovered signal transduction pathways. This would open a new area of investigation for these ubiquitious crystallins. Finally, we have completed the publication examining by differential display the pattern of expressed genes in epithelia from normal and cataractous human lenses reported initially last fiscal year. The results revealed 3 mRNAs with higher expression and 12 with lower expression in the cataractous lenses. Protein phosphatase 2A regulatory subunit mRNA, a mitoic suppressor, was lowered, while metallotheonein IIa, a detoxification protein, was increased. Studies on jellyfish and scallops, both species with remarkably sophisticated eyes with cellular lenses, have continued with the goal of understanding the evolutionary origins of vertebrate and ultimately human eyes. This year our publication, reported last fiscal year, appeared showing that jellyfish have an RXR gene whose product is expressed during development, binds 9-cis retinoic acid and can bind to the jellyfish J1-crystallin promoters. This is the first demonstration of an RXR in diploblastic species and suggests that regulation of crystallin gene expression by retinoid signalling may extend to the invertebrates. This year we have also extended our studies showing that the scallop lens have one major crystallin closely related to omega-crystallin/aldehyde dehydrogenase 1/2 of cephalopods. Although it appears to have the active site by sequence analysis, it lacks enzymatic activity with all substrates tested. When purified it exists as a dimer. Immunoflouresce tests showed that the scallop omega-crystallin/aldehyde dehydrogenase 1/2 is present a lower concentrations in the single epithelial cell layer comprising the scallop cornea. The isolation of its gene is in progress. The presence of this enzyme-related crystallin in scallops extends the existence of enzyme-crystallins to these invertebrates. Finally, we have identified an abundant 22 k Daprote in (called O-rystallin) comprising 3-5% of the soluble protein of the octopus lens. O- crystallin is 35-43% identical to a family of phosphatidylethanolamine-binding proteinsof vertebrates.